DESCRIPTION: Using individual rat tumors in vivo, a detailed analysis will be made of the relationship between radiation therapy resistance and newly-developed diagnostic assays for hypoxia, thiols, cell respiration and tumor blood flow. Hypoxia and thiols are known to increase the radiation resistance of cells in vitro, independently of all other known causes of therapy resistance. Our central hypothesis is that hypoxia and thiols act in a synergistic manner to increase the radiation resistance of tumor cells in vivo. Thus, measurements of these chemical factors should predict the radiation resistance of individual tumors. Various specific hypotheses will be tested: I) Reliable estimates of radiation therapy resistance depend directly on the thiol concentration and inversely on the oxygen concentration. Oxygen concentrations will be assayed by measuring metabolism- produced adducts of EF5, a 2-nitroimidazole, using monoclonal antibodies. Glutathione and cysteine concentrations will be assayed by HPLC with electrochemical detection. Both assays will use clinically relevant biopsies. Radiation response will be measured by irradiating tumors in vivo, then plating cells from the tumors for clonogenicity. II) Low oxygen concentrations are a direct result of diminished blood flow. Blood-flow measurements will be made at a microscopic and macroscopic level. The former will be assessed by the distribution of Hoechst 33342 dye and other methods. The latter will be assessed by direct flow and laser and ultrasound Doppler techniques. III) Non-invasive ultrasound Doppler imaging can select areas of tumors most likely to contain hypoxic cells. Although Doppler-based methods cannot provide absolute measures of blood flow they may be able to select for regions of tumors with relatively low flow. This would enable a new biopsy strategy to minimize problems caused by intratumoral heterogeneity of the distribution of hypoxia. IV) Non-invasive methods for imaging can provide quantitative analysis of tissue hypoxia. The 2-nitroimidazole drugs used in our current studies can be used for magnetic resonance spectroscopy and imaging (MRS/MRI). The biodistribution and tumor:tissue ratios will be studied as a function of drug concentration. Verification of spectroscopy will be made by comparing the NMR values with the actual drug-adduct distribution determined by antibodies and immunohistochemistry.